Generalized TSKF-Based Aeroengine Actuator Fault Estimation under Multi-source Uncertainty

The aeroengine epistemic uncertainty and the control system stochastic uncertainty form the multi-source uncertain environment during engine operation. On the basis of polynomial chaos expansion (PCE), the statistical property quantification of the discrete system matrices with probabilistic parameters is realized. Meanwhile, optimal filtering under multisource uncertainty is also achieved. And the generalized twostage Kalman filter is also proposed to achieve optimal filtering under multi-source uncertain environment. By treating actuator fault as unknown input, the estimation of actuator fault is provided based on the sub-filter. Numerical simulation shows that the proposed generalized TSKF is efficient for aeroengine actuator fault estimation under multi-source uncertainty.